ABSTRACT Our previous randomized clinical trials evaluated CLOSED LOOP insulin delivery in over 120 subjects under FREE-LIVING UNSUPERVISED HOME settings in studies lasting 1 WEEK to 3 MONTHS (total closed loop operation 81,120 hours/3,380days/9.3 years). All studies used CAMBRIDGE model predictive control algorithm. We demonstrated the safety, efficacy, and acceptability of closed loop insulin delivery over short to medium long investigations. Given its potential to reduce the BURDEN OF HYPOGLYCAEMIA and diabetes SELF-MANAGEMENT, application of closed-loop insulin delivery is of interest in older subjects due to the increased risk of SEVERE HYPOGLYCAEMIA in this population. The present project addresses the important question of FREE-LIVING UNSUPERVISED HOME USE of closed loop by a system configured to PROTECT against HYPOGLYCEMIA in OLDER PEOPLE with type 1 diabetes, by applying ADVANCED Android-phone closed loop system with pump-integrated HYPOGLYCAEMIA safety feature (predictive low glucose suspend to reduce the risk of hypoglycemia in case of Android-phone directed closed loop not being operational). Specifically, in the TEACHNICAL part of the project we will organize FOCUS GROUP meetings with older subjects from participating clinical sites. We will assess what specific closed loop FEATURES are required for this population. The additional feature(s) including personalized TARGET GLUCOSE SETTINGS will be IMPLEMENTED, TESTED and OPTIMIZED on Cambridge in silico population. REGULATORY DOCUMENTATION will be prepared. In the CLINICAL part of the project, we will evaluate the performance of 4 MONTH 24/7 CLOSED LOOP in OLDER SUBJECTS aged 60 years with type 1 diabetes under FREE-LIVING UNSUPERVISED HOME SETTINGS in a RANDOMISED CROSS-OVER design study. We will use closed loop system comprising 640G PUMP and ENLITE 3 sensor (both Medtronic) and an Android phone running CAMBRIDGE MODEL PREDICTIVE CONTROL ALGORITHM. The system SAFETY will be ENHANCED by the use of predictive low glucose management, INTEGRATED on 640G pump, and suspending insulin delivery when hypoglycemia is predicted even when Android-driven closed loop is not operating. In total, 36 PARTICIPANTS will be randomized to 4 month 24/7 CLOSED LOOP followed by SENSOR AUGMENTED PUMP THERAPY, or vice versa, with a 8-week washout period in between the two interventions. The study will compare BIOMEDICAL, PSYCHOSOCIAL, CARDIAC ARRYTHMIA EVENTS and SLEEP QUALITY of the two study interventions. Time spent within GLUCOSE TARGET RANGE measured by sensor glucose will be the primary endpoint. Biomedical outcomes (HbA1c, mean glucose, time below/above target and other related glucose/insulin metrics) will be complemented by assessments of psychosocial/human factors issues (focus groups and questionnaires), CARDIAC ARRHYTHMIA (ambulatory Holter) and SLEEP QUALITY (Actiwatch).